[rds-devel] testing with RDS on the mainline kernel

Or Gerlitz ogerlitz at voltaire.com
Thu May 15 07:58:26 PDT 2008


Olaf Kirch wrote:
> Indeed, it seems it's no longer exported in 2.6.26 :-( I guess all you
> can do here is do a copy-and-paste and send email to netdev.
I see now this commit, there's no way lkml/netdev would want to hear on 
problems related to non merged driver! Can you spare few words on why 
rds needs move_addr_to_kernel() is it some optimization to avoid calling 
kmap()?

> commit cd58f2a96ba95fb5b69580784bc6f7179001869c
> Author: Adrian Bunk <bunk at kernel.org>
> Date:   Wed Apr 23 03:37:49 2008 -0700
>
>     net: Unexport move_addr_to_{kernel,user}
>     
>     After the removal of the Solaris binary emulation the exports of
>     move_addr_to_{kernel,user} are no longer used.
>     
>     Signed-off-by: Adrian Bunk <bunk at kernel.org>
>     Signed-off-by: David S. Miller <davem at davemloft.net>
>
> diff --git a/net/socket.c b/net/socket.c
> index 9b5c917..66c4a8c 100644
> --- a/net/socket.c
> +++ b/net/socket.c
> @@ -2327,9 +2327,6 @@ int kernel_sock_shutdown(struct socket *sock, 
> enum sock_shutdown_cmd how)
>      return sock->ops->shutdown(sock, how);
>  }
>  
> -/* ABI emulation layers need these two */
> -EXPORT_SYMBOL(move_addr_to_kernel);
> -EXPORT_SYMBOL(move_addr_to_user);
>  EXPORT_SYMBOL(sock_create);
>  EXPORT_SYMBOL(sock_create_kern);
>  EXPORT_SYMBOL(sock_create_lite);
>
>
>
> It seems sk_alloc also needs this &init_net pointer.


thanks

Or.
> Olaf





More information about the rds-devel mailing list